U.S. patent application number 14/703038 was filed with the patent office on 2016-07-21 for automated cancellation of harmonics using feed forward filter reflection for radar transmitter linearization.
The applicant listed for this patent is U.S. Army Research Laboratory ATTN: RDRL-LOC-I. Invention is credited to Kyle A. Gallagher, Gregory J. Mazzaro.
Application Number | 20160209494 14/703038 |
Document ID | / |
Family ID | 56407698 |
Filed Date | 2016-07-21 |
United States Patent
Application |
20160209494 |
Kind Code |
A1 |
Mazzaro; Gregory J. ; et
al. |
July 21, 2016 |
AUTOMATED CANCELLATION OF HARMONICS USING FEED FORWARD FILTER
REFLECTION FOR RADAR TRANSMITTER LINEARIZATION
Abstract
A novel methodology for automatically linearizing a harmonic
radar transmitter--termed Feed-Forward Filter Reflection (FFFR)--is
disclosed. The method combines the reflected second harmonic from a
filter with the signal passing directly through the filter. The
second harmonic from these two paths are combined with equal and
opposite amplitudes to reduce the second harmonic beyond filtering
alone. This methodology has been experimentally verified at
transmit frequencies between 800 and 1000 MHz. Implemented
properly, the technique provides greater than 100 dB rejection
between 1.6 and 2.0 GHz. Although the tuning has been automated,
further optimization is possible. Automated tuning is demonstrated
over 400 MHz of bandwidth with a minimum cancellation of 110 dB.
One application for the harmonic cancellation is to create a linear
radar transmitter for the remote detection of non-linear
targets.
Inventors: |
Mazzaro; Gregory J.;
(Charleston, SC) ; Gallagher; Kyle A.; (Derwood,
MD) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
U.S. Army Research Laboratory ATTN: RDRL-LOC-I |
Adelphi |
MD |
US |
|
|
Family ID: |
56407698 |
Appl. No.: |
14/703038 |
Filed: |
May 4, 2015 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61987812 |
May 2, 2014 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G01S 7/038 20130101;
G01S 7/35 20130101; G01S 7/282 20130101; G01S 7/032 20130101 |
International
Class: |
G01S 7/40 20060101
G01S007/40 |
Goverment Interests
GOVERNMENT SPONSORSHIP
[0002] This invention was made with government support under Grant
No. W911NF-12-1-0305, awarded by the U.S. Army Research Office. The
Government has certain rights in the invention.
Claims
1. A system for linearizing a harmonic radar transmitter including
a power amplifier outputting a transmit signal with a transmit
frequency and an amplifier-generated harmonic, the system
comprising: a directional coupler in series with an RF filter and a
signal combiner; wherein the transmit signal from the power
amplifier passes through the directional coupler to the RF filter;
the RF filter configured to reject the amplifier-generated
harmonic, such that the amplifier-generated harmonic is reflected
back through the coupler in the reverse direction, and the filter
passes the transmit frequency and system-generated harmonic to the
signal combiner; wherein the reflected amplifier-generated harmonic
from the coupler passes through a variable attenuator and a phase
shifter, the variable attenuator configured to reduce the magnitude
of the reflected amplifier-generated harmonic to match that of the
system-generated harmonic, and the phase shifter configured to
cause the reflected amplifier-generated harmonic to be 180 degrees
out of phase with the system-generated harmonic, such that the
amplifier-generated and system-generated harmonics cancel out at
the combiner; and the combiner outputs the transmit frequency
without the harmonic to a transmit antenna.
2. The system of claim 1, wherein the variable attenuator and phase
shifter are automatically adjusted so that the amplifier-generated
and system-generated harmonics cancel out at the combiner.
3. The system of claim 1, further comprising a radar receiver
including a receive antenna coupled to a filter configured to
reject the transmit frequency and pass the harmonic.
4. The system of claim 1, further including a fixed attenuator
configured to shift the cancellation frequency to cancel different
frequencies.
5. The system of claim 1, further including different couplers
configured to tune out lower frequencies requiring less
attenuation
6. The system of claim 1, further including an amplifier in series
with the phase shifter configured to cancel lower frequencies
having less loss through the filter.
7. The system of claim 1, further including two vector modulators
and an amplifier configured to improve the frequency tuning range
on the high and low sides.
8. A system for linearizing a harmonic radar receiver, the system
comprising: an antenna configured to receive a fundamental transmit
frequency and a desired harmonic; a high pass filter coupled to the
antenna that is configured to pass the harmonic and reject the
fundamental; a directional coupler configured to receive the
rejected fundamental and redirect it to a variable attenuator and a
phase shifter; and one or more vector modulators configured to
control the variable attenuator and a phase shifter to cancel the
fundamental transmit frequency via summation with the rejected
fundamental.
Description
RELATED APPLICATION(S)
[0001] This application claims the benefit of U.S. Provisional
Patent Application No. 61/987,812 filed May 2, 2014, herein
incorporated by reference in its entirety for all purposes.
BACKGROUND OF THE INVENTION
[0003] 1. Field
[0004] Embodiments of the present invention are generally directed
to radar, and more particularly, to a methodology for automated
cancellation of harmonics using feed forward filter reflection for
radar transmitted linearization.
[0005] 2. Description of Related Art
[0006] Harmonic radar exploits harmonically generated returns from
nonlinear targets to aid in their detection. The advantage of
nonlinear radar over traditional radar is its high clutter
rejection, as most naturally-occurring (clutter) materials do not
exhibit a nonlinear electromagnetic response under illumination by
radio-frequency (RF) energy. The disadvantage of nonlinear radar is
that the power-on-target required to generate a signal-to-noise
ratio (SNR) comparable to linear radar is much higher than that of
linear radar. Nevertheless, nonlinear radar is particularly suited
to the detection of man-made electronic devices, typically those
containing semiconductors whose radar cross-section is very low
owing to their thin geometric profile.
[0007] A nonlinear radar tailored to a set of RF electronic
responses would help law enforcement agents locate devices whose
emissions exceed those permitted by law, allow security personnel
to detect unauthorized radio electronics in restricted areas, or
enable first-responders to pinpoint personal electronics during
emergencies such as immediately after an avalanche or
earthquake.
[0008] Harmonic radar is a type of nonlinear radar that transmits a
single frequency f.sub.0 and receives one or more integral
multiples of that same frequency (e.g. 2f.sub.0, 3f.sub.0,
4f.sub.0, etc.). The most common harmonic radars receive the lowest
harmonic, 2f.sub.0, because 2f.sub.0 tends to be the strongest of
all harmonics generated by an electronic target for a given
transmit frequency and power.
[0009] In order to generate a detectable harmonic response from an
electronic device, the required power density on dBmW target is
approximately
10 dBm W cm 2 , ##EQU00001##
which is comparable to the power density observed directly below a
cellular base station. Thus, the harmonic radar's transmitter must
provide high power to overcome interference by possible cellular
towers in the vicinity. Also, typical harmonic responses are
received power levels of -100 dBm, and this weak signal must not be
masked by harmonics generated by the transmitter that are coupled
directly to the receiver. Thus, the harmonic radar's high transmit
power must be provided with high linearity.
Basic Harmonic Radar
[0010] A simple harmonic radar is shown in FIG. 1A. The transmitter
consists of a synthesizer which outputs a single frequency f.sub.0,
a power amplifier (PA) which boosts the transmit signal to a level
suitable for exciting a harmonic response from the target, and a
transmit (T.sub.x) antenna. The single frequency f.sub.0
illuminates the target and the harmonic 2f.sub.0 radiates from the
target back towards the radar. The receiver consists of a receive
(R.sub.x) antenna, a low-noise amplifier for boosting the received
signal to a level suitable for capture, and an analog-to-digital
converter (ADC) which records the received signal. Target detection
may be performed using this continuous-wave configuration. Ranging
may be accomplished by pulsing or otherwise modulating the
transmission.
[0011] Unfortunately, practical harmonic radar design is not so
straightforward, as illustrated in FIG. 1B. Three reasons
highlighted below are:
[0012] The power amplifier, in addition to boosting the transmitted
tone, generates harmonics. If the amplifier-generated harmonic
2f.sub.0 is not attenuated sufficiently before it arrives at the
T.sub.x antenna, this harmonic will be radiated from the
transmitter, reflect from the target, and mask the target's
harmonic response.
[0013] In any practical radar system (linear or nonlinear),
coupling exists between the transmitter and the receiver. If this
coupling is excessively high, the transmitter-generated 2f.sub.0
will be fed directly to the receiver and will mask the target's
harmonic response.
[0014] The target's linear response (at f.sub.0, reflected from the
target casing) will likely be much stronger than its nonlinear
response (at 2f.sub.0, radiated from the target electronics). Thus,
even if the T.sub.x/R.sub.x antenna coupling is minimal, a strong
signal at f.sub.0 will enter the receiver at the R.sub.x antenna.
If this received f.sub.0 is not attenuated sufficiently before the
low noise amplifier (LNA), the LNA will pass f.sub.0 to the ADC
(possibly saturating the converter) and/or it will produce its own
2f.sub.0 to mask the target's harmonic response.
[0015] For these highlighted reasons, reduction of the
system-generated harmonics, i.e. "linearization" of the radar, is
necessary.
Linearized Harmonic Radar
[0016] Two popular techniques for RF linearization are filtering
and feed-forward cancellation. Filtering removes system-generated
harmonics by attenuating or reflecting them at the output of the
nonlinearity. Feed-forward cancellation adds a phase-shifted
version of the undesired signal to the combined signal in order to
remove the undesired signal. The undesired signal may be a
harmonic, or it may be a strong linear signal that is likely to
generate a harmonic.
[0017] Filtering may be implemented in the transmitter and/or the
receiver. It is depicted in FIG. 2 as part of the transmitter
identified as (A). Here, a lowpass filter removes 2f.sub.0 at the
output of the amplifier, which prevents the amplifier-generated
2f.sub.0 from radiating out of the T.sub.xantenna. Cancellation is
usually implemented in the receiver. It is depicted in FIG. 2
identified as (B) inserted at the junction between the R.sub.x
antenna and the LNA. A 180.degree. phase-shifted version of f.sub.0
is added to the signal received from the R.sub.x antenna, where the
signal includes the target response at f.sub.0 as well as 2f.sub.0.
The vector sum of the phase-shifted f.sub.0 with the un-shifted
f.sub.0 is ideally zero, ensuring that only 2f.sub.0 appears at the
output of the cancellation circuit and continues along the receiver
chain.
[0018] The architecture presented in FIG. 2 is still not an
adequate nonlinear target detector for practical standoff ranges
between the T.sub.x/R.sub.x antennas and an electronic target. A
typical electronic target response, at a distance of 3 m,
illuminated by 1 W at 800 MHz from a T.sub.x antenna with a gain of
9 dBi, and received at 1600 MHz from an R.sub.x antenna with a gain
of 10 dBi, is approximately P.sub.target=-90 dBm. For 10 W transmit
power and a distance of 20 m, the target response drops to
P.sub.target=-130 dBm. Assume that P.sub.trans=40 dBm at f.sub.0
and that the coupling directly between P.sub.trans and P.sub.rec
(at all frequencies, for simplicity) is .DELTA.P.sub.coupled=30 dB.
If the amplifier generates P.sub.trans=-30 dBm at 2f.sub.0 and if
the filter is capable of rejecting 2f.sub.0 at the output of the
amplifier by 60 dB, then the transmitter-generated harmonic that
couples directly to the receiver is
P system ( 2 f 0 ) = P tran . ( 2 f 0 ) - S 21 filter ( 2 f 0 ) -
.DELTA. P coupled ( 2 f 0 ) = - 30 dBm - 60 dB - 30 dB = 120 dBm (
1 ) ##EQU00002##
which is above the target response (P.sub.system>P.sub.target)
by 10 dB, for a distance of 20 m. This scenario is illustrated in
FIG. 3, for a finite frequency band f.sub..alpha. to f.sub..beta.
over which the target emits a measurable harmonic response. As
depicted, the target response at 2f.sub.0 is masked by the
system-generated harmonic distortion at 2f.sub.0; thus, the target
cannot, in theory, be detected.
[0019] To reduce the system-generated distortion P.sub.system to a
level below the target response P.sub.target, several approaches
can be taken:
[0020] The rejection provided by the lowpass filter can be
increased. The tradeoff is increased passband loss. The signal will
need to traverse additional filter elements that will not only
increase rejection in the stopband but will also increase loss in
the passband. Also, lowpass filtering is a fixed solution that does
not allow tuning the circuit to reject particular harmonics (e.g.
if f.sub.0 and 2f.sub.0 are not known).
[0021] A bandstop filter can be substituted for the lowpass filter.
The tradeoff is a periodic pattern in frequency for the passband,
which may be undesirable if the receiver is sensitive to
high-frequency noise and/or higher harmonics. Like the lowpass
filter, the bandstop filter is a fixed-frequency solution.
[0022] A tunable lowpass filter may be substituted for the fixed
lowpass filter. This solution is still not ideal because tuning is
typically accomplished mechanically, which is slow compared to the
change-of-frequency required for a practical radar system such as a
stepped-frequency radar, electronically, which degrades the
linearity of the transmitted signal.
[0023] A filter bank may be implemented with electronic switches.
However, insertion of switches to select between multiple filters
will increase the loss of the overall filtering structure, and/or
degrade the linearity of the transmitted signal.
[0024] Further improvements would be useful.
BRIEF SUMMARY OF THE INVENTION
[0025] Embodiments of the present invention are generally directed
to a methodology for automated cancellation of harmonics using feed
forward filter reflection for radar transmitted linearization.
[0026] According to embodiments, a system is provided for
linearizing a harmonic radar transmitter that includes a power
amplifier outputting a transmit signal with a transmit frequency
and an amplifier-generated harmonic. More particularly, the system
comprises a directional coupler in series with an RF filter and a
signal combiner with the transmit signal from the power amplifier
passes through the directional coupler to the RF filter. The RF
filter is configured to reject the amplifier-generated harmonic,
such that the amplifier-generated harmonic is reflected back
through the coupler in the reverse direction, and the filter passes
the transmit frequency and system-generated harmonic to the signal
combiner. The reflected amplifier-generated harmonic from the
coupler passes through a variable attenuator and a phase shifter,
with the variable attenuator configured to reduce the magnitude of
the reflected amplifier-generated harmonic to match that of the
system-generated harmonic, and the phase shifter configured to
cause the reflected amplifier-generated harmonic to be 180 degrees
out of phase with the system-generated harmonic, such that the
amplifier-generated and system-generated harmonics cancel out at
the combiner. The combiner then outputs the transmit frequency
without the harmonic to a transmit antenna.
[0027] The variable attenuator and phase shifter can be
automatically adjusted so that the amplifier-generated and
system-generated harmonics cancel out at the combiner. In some
embodiments, the system may further include: (i) a radar receiver
including a receive antenna coupled configured to a filter to
reject the transmit frequency and pass the harmonic; (ii) a fixed
attenuator configured to shift the cancellation frequency to cancel
different frequencies; (iii) different couplers configured to tune
out lower frequencies requiring less attenuation; (iv) an amplifier
in series with the phase shifter to configured cancel lower
frequencies having less loss through the filter; and/or (v) two
vector modulators and an amplifier configured to improve the
frequency tuning range on the high and low sides.
[0028] According to other embodiments, a system for linearizing a
harmonic radar receiver includes: an antenna configured to receive
a fundamental transmit frequency and a desired harmonic; a high
pass filter coupled to the antenna that is configured to pass the
harmonic and reject the fundamental; a directional coupler
configured to receive the rejected fundamental and redirect it to a
variable attenuator and a phase shifter; and one or more vector
modulators configured to control the variable attenuator and a
phase shifter to cancel the fundamental transmit frequency via
summation with the rejected fundamental.
[0029] These and other embodiments of the invention are described
in more detail, below.
BRIEF DESCRIPTION OF THE DRAWINGS
[0030] So that the manner in which the above recited features of
the present invention can be understood in detail, a more
particular description of the invention, briefly summarized above,
may be had by reference to embodiments, some of which are
illustrated in the appended drawings. It is to be noted, however,
that the appended drawings illustrate only typical embodiments of
this invention and are therefore not to be considered limiting of
its scope, for the invention may admit to other equally effective
embodiments, including less effective but also less expensive
embodiments which for some applications may be preferred when funds
are limited. These embodiments are intended to be included within
the following description and protected by the accompanying
claims.
[0031] FIG. 1 shows a simple harmonic radar, where FIG. 1A shows
its ideal performance and FIG. 1B show its actual performance.
[0032] FIG. 2 show a harmonic radar linearized by filtering and
feed-forward cancellation.
[0033] FIG. 3 shows transmitted and received spectra for
P.sub.system>P.sub.target showing insufficient transmitter
linearization.
[0034] FIG. 4 shows a schematic for harmonic radar with transmitter
linearization by feed-forward filter reflection according to
embodiments of the present invention.
[0035] FIG. 5 is a simple illustration showing how the phase
changes through the different paths in a feed-forward filter
reflection system.
[0036] FIG. 6 shows filter response of the feed-forward filter
reflection in one exemplary embodiment.
[0037] FIG. 7 shows a feed-forward filter reflection implementation
according to an exemplary embodiment.
[0038] FIG. 8 is a plot of frequency response of feed-forward
filter reflection tuning out various frequencies.
[0039] FIG. 9 is a plot of measured optimum vector modulation (VM)
phase data along with a best-fit line.
DETAILED DESCRIPTION OF THE INVENTION
[0040] This invention resides in a system and methods for automatic
linearizing a harmonic radar transmitter. In the preferred
embodiments, a transmitter for a harmonic radar transmits f.sub.0
and receives only 2f.sub.0. The solution involves a form of tunable
lowpass filter whose passband remains unaltered by tuning but whose
deepest stopband rejection (notch) frequency is flexible. The notch
frequency corresponds to the harmonic to be received from the
target. For this study, the notch is tuned to 2f.sub.0, but it may
also be tuned to a higher harmonic (e.g. 3f.sub.0, 4f.sub.0, etc.)
whose reception is used for target classification.
[0041] Tuning occurs in a low-power branch parallel to the
lowpass-filtered signal, which: minimizes signal loss, enables
fast, electronic tuning, maintains linearity in the high-power
branch. This invention demonstrates the feasibility of
electronically tuning such a system.
Feed-Forward Filter Reflection (FFFR)
[0042] A novel linearization methodology for harmonic radar is
presented in FIG. 4. The linearization circuit consists of a
filter, a directional coupler, an amplitude modulator, a phase
shifter, and a combiner (summation circuit). From FIG. 4, the power
amplifier feeds Port A with both the desired frequency f.sub.0 and
the undesired harmonic. The transmit signal propagates through the
coupler with minimal loss and arrives at the filter. The filter is
designed to pass the desired frequency and to reject the
amplifier-generated harmonic (in the forward direction) according
to its two-port transmission characteristic |S.sub.21|. At the
output of the filter, the transmitted signal is partially
linearized, but still contains enough of the system-generated
harmonic to mask the target response.
[0043] RF filters are generally designed to reflect (and not
attenuate) the frequencies that they reject. Thus, the rejected
harmonic reflects from the filter and propagates in the reverse
direction, back through the directional coupler. The coupler
captures some of the reverse-traveling wave (typically 10 dB or 20
dB) and sends the signal to a variable attenuator and phase
shifters. This pair modifies the harmonic to a wave that is equal
in magnitude and opposite in phase to the harmonic output from the
filter. We will call this the cancellation path. When the filtered
output is summed with the cancellation path the second harmonic
cancels and the signal, P.sub.trans, is fully linearized.
Automating Tuning of FFFR Circuit
[0044] As stated in the previous sections, in order to cancel the
second harmonic the phase difference between the filter and
cancellation paths needs to be exactly 180 degrees at the desired
frequency. The magnitude of the two signals also needs to be equal
at the time on recombination. Thus, it is natural to break up the
tuning process into tuning the phase to be 180 degrees out of phase
and tuning the magnitude to be the same.
Tuning Phase
[0045] A simple illustration of how the phase changes through the
different paths is shown in FIG. 5. In FIG. 5, R.sub.f and R.sub.c
are the electrical lengths of the filter and cancellation paths
respectively, .beta. is the wave number, .phi..sub.m is the phase
change caused by the vector modulator (VM) and .phi.f.sub.p and
.phi..sub.cp are the total changes is phase in the filter path and
cancellation path, respectively. To stay general, the phase
difference between the cancellation path (.phi..sub.cp) and filter
path (.phi.f.sub.p) could be 180.degree.+n.times.360.degree., where
n=0, 1, 2. This condition for cancellation can be written in
radians as shown in Eq. 2. Also, .phi.f.sub.p and .phi..sub.cp are
defined in Eq. 3.
.phi..sub.fp-.phi..sub.cp=(2n-1).pi.,n=1,2,3 (2)
.phi..sub.fp=.beta.R.sub.f,.phi..sub.cp=.beta.R.sub.c+.phi..sub.m
(3)
Plugging Eq. 3 into Eq. 2, we obtain
.beta.R.sub.f-(.beta.R.sub.c+.phi..sub.m)=(2n-1).pi. (4)
[0046] Rearranging terms in Eq. 4, it is easy to get the expression
in Eq. 5
.beta.(R.sub.f-R.sub.c)-.phi..sub.m=(2n-1).pi. (5)
[0047] The substitutions made to get from Eq. 4 to Eq. 5 assume
that the speed of the wave in both the filter path and cancellation
path are the same. This is not an unrealistic assumption when
tuning out single frequencies at a time, but if tuning is to be
done across a wide bandwidth at the same time, dispersion needs to
be taken into account.
[0048] To further reduce the condition for cancellation two
addition substitutions are made, namely, R.sub.f-R.sub.c=.DELTA.R
and .beta.=2.pi.f/.nu., where .nu. is the speed of the wave inside
the circuit. After making the substitutions and solving for f, the
tuning frequency, the tuning expression can be expressed as in Eq.
6.
f = v 2 .pi. .DELTA. R .PHI. m + ( 2 n - 1 ) v 2 .DELTA. R ( 6 )
##EQU00003##
Tuning the Magnitude
[0049] The above equations show how changing the vector modulator
phase changes the cancellation frequency. The other half of the
tuning process is matching the loss through the filter and
cancellation paths. The loss through the filter path is set by the
LPF. The loss through the filter path can be expresses as the
transfer function of the filter, or |S.sup.f.sub.21|. The loss
through the cancellation path is set by the two couplers and the
loss through the VM. Each coupler has 20 dB of loss and the VM can
change its loss from -5 dB to -35 dB. Therefore, the total loss
through the cancellation path is between 45 and 75 dB. This means
that the cancellation path can cancel any frequency that has
between 45 and 75 dB of loss through the filter path. For the
Mini-Circuits NLP-1000+ filter, this means that any frequency
between 1.57 and 2.35 GHz can be canceled, as seen in FIG. 6.
[0050] If different frequencies need to be canceled, a fixed
attenuator can be added to shift the cancellation frequency higher.
To tune out lower frequencies, less attenuation is needed so
different couplers can be used, say -10 dB in place of the -20 dB.
An amplifier can also be added to cancellation path to cancel lower
frequencies that have less loss through the filter. To improve the
frequency tuning range on the high and low side, two VMs can be
used in addition to an amplifier. For our application, canceling
frequencies between 1.6-2.0 GHz, one VM and no amplifiers are
needed.
Implemention of Automated FFFR Tuning
[0051] The VM is controlled with the analog outputs on a NI DAQ
data acquisition system using LabView and an Agilent PNA N5225A
network analyzer is also controlled with LabView. The PNA is setup
in S.sub.21 mode and is connected to Ports A and B of the FFFR
network, shown in FIG. 4. The PNA provides the feedback about how
well the FFFR is canceling the desired frequency. The frequency
required to cancel is inputted into LabView and the program first
sweeps the phase of the VM. Sweeping the phase corresponds to
changing the frequency of the notch created from the cancellation,
from Eq. 6. Data from the PNA at the desired frequency are stored
as the phase of the VM is swept. The data are stored as S.sub.21 at
each frequency vs VM phase. The minimum of the S.sub.21 data is
found. The phase that provides the minimum S.sub.21 corresponds to
the phase required to cancel the desired frequency.
[0052] Once the desired phase is found, the magnitude of the VM is
swept. S.sub.21 measurements are taken at the desired frequency as
the VM sweeps magnitude. A plot of S.sub.21 vs. VM magnitude is
generated and the minimum is found. The VM magnitude of the minimum
of the S.sub.21 corresponds to the amount of attenuation needed to
match the filter path. The sweeping of the magnitude and phase is
done several times to achieve an S.sub.21 value of less than 110
dB.
[0053] With no a priori knowledge of the magnitude or phase
required to cancel the desired frequency this technique will go
through 100-120 combinations of magnitude and phase before reaching
the -110 dB S.sub.21 desired cancellation. However, with knowledge
of the required magnitude and phase, only 10-20 combinations are
needed. It would be apparent to a person of skill in the art that
optimization techniques can be implemented to reduce the number of
iterations.
Measurements and Results
[0054] The implementation of the automated FFFR technique has been
chosen for a harmonic radar to operate over the frequency range of
800-1,000 MHz. Therefore, the second harmonic will be generated
from 1,600-2,000 MHz. For this reason the frequency range chosen
for the frequency rejection is 1,600-2,000 MHz, with the LPF
passing DC--1,000 MHz.
[0055] The FFFR technique has been implemented using the RF circuit
shown in FIG. 7. The directional coupler is the HP 778D, the
variable attenuator and phase shifter are implemented with an
Analog Devices (AD8341), Vector Modulator (VM), and another
combiner HP 778D. The lowpass filter tested is the Mini-Circuits
NLP-1000+.
[0056] The frequency response from Port A to Port B is measured
with the VM tuned to cancel 5 frequencies, namely, 1.6, 1.7, 1.8,
1.9 and 2.0 GHz. As stated, this frequency range corresponds to the
second harmonic that would be generated by a power amplifier
transmitting between 800 and 1,000 MHz. The Agilent PNA is used to
collect the frequency response, S.sub.21. FIG. 8 shows the
results.
[0057] Each trace corresponds to the FFFR tuning out a different
frequency. In addition to storing the S.sub.21 data from each
tuning, the magnitude and phase settings of the VM are recorded.
They are given in Table I.
TABLE-US-00001 TABLE I Vector Modulator (VM) settings for canceling
1.6, 1.7, 1.8, 1.9 and 2.0 GHz Tuning Frequency (GHz) Magnitude of
VM Phase of VM (Rad./.pi.) 1.6 0.72 1.55 1.7 0.44 1.90 1.8 0.33
2.16 1.9 0.16 2.42 2.0 0.085 2.86
[0058] In an effort to speed up the automation process, the VM
phase is compared to the tuning frequency and a line of best fit is
obtained. This line of best fit relates the desired frequency to
cancel to the required VM phase setting. This equation follows from
Eq. 6. A plot with the measured data and the fitted line is shown
in FIG. 9.
[0059] The equation for the fitted line is:
f ( MHz ) = 304.2 ( MHz rad ) .times. .PHI. m ( rad ) + 1 , 129 (
MHz ) ( 7 ) ##EQU00004##
[0060] Equation 7 was developed from the cancellation condition,
Eq. 3, and directly yields the cancellation frequency for a given
phase. A more practical manipulation of the equations yields the
required tuning phase to cancel a desired frequency, as given in
Eq. 8.
.PHI. m ( rad ) = f ( MHz ) - 1 , 129 ( MHz ) 304.2 ( MHz / rad ) (
8 ) ##EQU00005##
[0061] Using Eq. 8, the required VM phase to cancel any frequency
within the tunable range can be found. The value of VM phase can
only serve as a starting point for tuning. If high cancellation is
required, 10-15 iterations of tuning are still required. The reason
for this is because if 100 dB of cancellation is required, the
phase difference between the two signal needs to be within 0.1% of
180.degree. and this level of accuracy is not obtainable with a
fitted line. This is because the tuning process is very sensitive
to small changes in cable length and temperature.
[0062] While the emphasis herein has been placed on the transmitter
side of the system, the FFFR technique can also be applied to the
receive side of a harmonic radar. A high pass filter can be used to
pass the harmonic and reject the fundamental. The fundamental
frequency signal can then be coupled off using the same couplers.
The magnitude and phase can be shifted using similar vector
modulators, depending on the frequency, and the same cancellation
should be achievable.
[0063] Aspects related to the invention have been previously
disclosed in: (1) K. A. Gallagher, G. J. Mazzara, K. D. Sherbondy,
R. M. Narayanan, and A. F. Martone, "Automated cancellation of
harmonics using feed-forward filter reflection for radar
transmitter linearization," Proceedings of the SPIE, Vol. 9077, pp.
90770U(1-10), May 2014 (presented at conference in Baltimore, Md.
on May 5, 2014; published in the SPIE proceedings on May 29, 2014);
and (2) K. Gallagher, R. Narayanan, G. Mazzara, K. Sherbondy,
"Linearization of a harmonic radar transmitter by feed-forward
filter reflection," Proceedings of the IEEE Radar Conference, May
2014. (presented in Cincinnati, Ohio on May 20, 2014), both
disclosures are herein incorporated by reference in their
entireties.
[0064] The foregoing description, for purpose of explanation, has
been described with reference to specific embodiments. However, the
illustrative discussions above are not intended to be exhaustive or
to limit the invention to the precise forms disclosed. Many
modifications and variations are possible in view of the above
teachings. The embodiments were chosen and described in order to
best explain the principles of the present disclosure and its
practical applications, and to describe the actual partial
implementation in the laboratory of the system which was assembled
using a combination of existing equipment and equipment that could
be readily obtained by the inventors, to thereby enable others
skilled in the art to best utilize the invention and various
embodiments with various modifications as may be suited to the
particular use contemplated.
[0065] While the foregoing is directed to embodiments of the
present invention, other and further embodiments of the invention
may be devised without departing from the basic scope thereof, and
the scope thereof is determined by the claims that follow.
* * * * *